This is something I preach often, even at FAANG. Often engineers feel comfortable and in-the-zone when solving for a process. One saying I often delivery is: Engineers often think their job is to build things. That's not the whole story, they build things for _people_. Nobody cares what your code looks like. Nobody cares what your architecture looks like. As engineers we should worry about creating forgettable experi…
Your tech stack is not the product
21–30 of 132 posts
Re: Your tech stack is not the product
#22This is something I preach often, even at FAANG. Often engineers feel comfortable and in-the-zone when solving for a process. One saying I often delivery is: Engineers often think their job is to build things. That's not the whole story, they build things for _people_. Nobody cares what your code looks like. Nobody cares what your architecture looks like. As engineers we should worry about creating forgettable experi…
Re: Your tech stack is not the product
#23> Why is Joe’s closet computer a bad choice? Because it’s a single point of failure and we won’t be able to ship fast if it breaks, which it will. It's kinda a shame they reduced it to this. A single machine in a colo center is going to be far more reliable than single availability zone in AWS, which is all many people resort to. And maybe we've just gotten lucky, but in general our very simple 20 machine colo center…
"It's kinda a shame they reduced it to this. A single machine in a colo center is going to be far more reliable than single availability zone" I think that depends on the colo honestly. What is so unreliable about a single EC2 instance in a zone?
Re: Your tech stack is not the product
#24Re: Your tech stack is not the product
#25This is something I preach often, even at FAANG. Often engineers feel comfortable and in-the-zone when solving for a process. One saying I often delivery is: Engineers often think their job is to build things. That's not the whole story, they build things for _people_. Nobody cares what your code looks like. Nobody cares what your architecture looks like. As engineers we should worry about creating forgettable experi…
Good architecture leads to faster feature development and increased stability, which makes your users happy
Re: Your tech stack is not the product
#26If you need to do something that's not just generic CRUD app/site... picking an exotic technology that high-powered programmers love, might make it easier to hire.
Compare:
"We're building a gig economy app to clean gas station bathrooms, and we eat our own dogfood!"
to:
"We're USING RUST to build something (not a crypto scam, honest), and WE WILL PAY YOU MONEY TO HACK RUST, and did we mention RUST!!!"
Re: Your tech stack is not the product
#27> Why is Joe’s closet computer a bad choice? Because it’s a single point of failure and we won’t be able to ship fast if it breaks, which it will. It's kinda a shame they reduced it to this. A single machine in a colo center is going to be far more reliable than single availability zone in AWS, which is all many people resort to. And maybe we've just gotten lucky, but in general our very simple 20 machine colo center…
"It's kinda a shame they reduced it to this. A single machine in a colo center is going to be far more reliable than single availability zone" I think that depends on the colo honestly. What is so unreliable about a single EC2 instance in a zone?
This isn't bias speaking, I work on Fly.io, our VMs are less reliable than EC2 VMs. AWS's pitch is that all the extra complexity in their infrastructure benefits you. So is ours! But it is, in fact, extra complexity that will bite you in the ass if you don't build your apps the right way.
Re: Your tech stack is not the product
#28> customers are not paying for, nor give a shit about, these things. Yes, but they do give a shit if: - your product is slow, so they assume it's crap and leave your site - your product breaks, no one notices or fixes it - your product is bad, because your wrote it in a safe boring technology like cobol and therefore couldn't access the best developers. (exaggeration, but variations on this are true) My point isn't t…
Yes, those things are always important (if not critical); it's not meant as a defense of bad architecture, merely to try to make those decisions through a product/customers-first lens.
(I did try to connect quality/reliablity, perhaps unsuccessfully, later in the post, e.g. in goals: "If the system is unreliable, we won’t ship as much product ... If we don’t replace this system, more and more customers will experience a broken product.")
Re: Your tech stack is not the product
#29> Why is Joe’s closet computer a bad choice? Because it’s a single point of failure and we won’t be able to ship fast if it breaks, which it will. It's kinda a shame they reduced it to this. A single machine in a colo center is going to be far more reliable than single availability zone in AWS, which is all many people resort to. And maybe we've just gotten lucky, but in general our very simple 20 machine colo center…
Colo is fine if you can set it up extremely quickly and it takes minimal ongoing support time. Depending on what you mean by Colo, that may or may not be possible.
Re: Your tech stack is not the product
#30Earlier quoted context omitted.
Forgive me if I'm wrong about this, but it would seem to me that the FAA NOTAM system had orders of magnitude more uptime than literally any product produced by so called "professional" "engineers" at big software companies.
AWS uptime is pretty insane